Constraint-based modelling using Genome-Scale Metabolic (GSM) models is useful in industrial biotechnology and systems medicine, where they are used to predict metabolic phenotypes, improve the productivity of microorganisms, elucidate mechanisms, predict biomarkers and drug targets of diseases, etc. Reactive Species (RS) like hydrogen peroxide, superoxide anion, nitric oxide radical, and hydroxyl radical are by-products of normal metabolic activities in living organisms and have notable roles during environmental stress induced states (acid, oxygen, shear, chemicals, antibiotics, etc.) in the case of prokaryotes and during infections and disease contexts in eukaryotes. Regardless of their ubiquitous roles in metabolism across all the domains of life (prokaryotes, archaea and eukaryotes), the RS are not adequately modelled in the GSM models. GSM models are used to understand cell metabolism in diverse contexts like microbial GSM models and established whole-body human reconstructions like Recon 3D and Human-GEM. A scalable Reactive Species Reactions Module (RSRM), which consists of 257 reactions and 310 metabolites, was developed that can be integrated with any metabolic model. Further, the usefulness of RSRM to improve GSM model predictions across two biological domains – a eukaryote (cancer cell) and a prokaryote (Sporosarcina pasteurii), are discussed as follows. Metabolic rewiring is a hallmark of cancer. RS levels are increased during cancer development, which causes redox imbalance. To highlight the relevant metabolic reprogramming, RSRM was integrated with the GSM models of three cancers, namely, Triple Negative Breast Cancer (TNBC), High-Grade Serous Ovarian Cancer (HGSOC) and Colorectal Cancer (CRC) constructed using cell line transcriptomics data from the Cancer Cell Line Encyclopedia database. With RSRM integration, the GSM models of these three cancers precisely highlighted the increase/decrease in fluxes (dysregulation) occurring in important pathways of these cancers through flux sampling analyses. These dysregulations in the metabolic pathways like glutamine metabolism and fatty acid metabolism, kynurenine pathway, glutathione metabolism, eicosanoid metabolism, serine metabolism, etc, were not prominent in the standard cancer models without the RSRM. The mechanism of oxidative stress induced killing of cancer cells by vitamin C application has been depicted using cancer GSM models. RSRM integration highlighted the distinct regulation of ferroptosis (a recently recognised form of regulated cell death that is characterised by lipid peroxidation and iron accumulation) across the three cancers. The results from these RSRM-integrated cancer GSM models suggested the following decreasing order in the ease of ferroptosis-targeting to treat the cancers: TNBC > HGSOC > CRC because TNBC and HGSOC have higher fluxes through ferroptosis promoting reactions compared to CRC. From the in-silico predictions, it has been shown that cysteine producing genes like CBS and CTH must be targeted for their roles in glutathione synthesis, and cystine import reaction mediated by SLC7A11 do not contribute to this process in cancers as stated in the literature. Now, moving on to the prokaryote, Microbially Induced Calcite Precipitation (MICP) or biocementation has applications in self-healing concrete, soil consolidation and bio-grouting. A GSM model of S. pasteurii, an aerobic bacteria used for biocementation in sub-surfaces, was built using omics data from literature and was constrained using experimental data. A prokaryotic RSRM tailored for S. pasteurii metabolite composition was built using the RSRM. The basal levels of RS in S. pasteurii were described by constraining the bounds of the reactions and then integrated to the GSM model of S. pasteurii. The prokaryotic RSRM comprises 115 reactions and 169 metabolites, and this basal RSRM integrated S. pasteurii GSM model was used for simulation studies on MICP. S. pasteurii is known to exhibit better growth and MICP activities during aerobic conditions than during anaerobic conditions. Studies have shown that hydrogen peroxide at a suitable concentration can be used as a Liquid-Phase Oxygen (LPO) supply during anaerobic sub-surface applications because of its ability to induce catalase in organisms. Catalase converts hydrogen peroxide into molecular oxygen, which gets used by organisms. Three condition-specific GSM models of S. pasteurii that describe the growth of S. pasteurii under different aeration conditions (oxygen, hydrogen peroxide as aeration source and anaerobic) were built and analysed. The catalase flux of the S. pasteurii GSM model was very high under LPO (hydrogen peroxide) condition. The urease flux and the biomass flux during LPO were comparable to the aerobic (oxygen) condition, and higher than anaerobic condition from the GSM model predictions. Further, the carbonic anhydrase flux necessary for calcite precipitation was found to be higher during LPO than under anaerobic condition. These results have shown that LPO could be used for aeration of S. pasteurii in oxygen-deficient conditions for MICP applications.
Keywords
Reactive species; Genome-Scale Metabolic (GSM) models; Cancer metabolism; Ferroptosis; Microbially Induced Calcite Precipitation (MICP); Ureolysis; Biocementation; Liquid-Phase Oxygen (LPO); Basal reactive species

Publications:
1. Sridhar, S., Bhatt, N., & Suraishkumar, G. K. (2021). Mechanistic insights into ureolysis mediated calcite precipitation. Biochemical Engineering Journal, 176, 108214. https://doi.org/10.1016/j.bej.2021.108214
2. Sridhar, S., Bhalla, P., Kullu, J., Veerapaneni, S., Sahoo, S., Bhatt, N., & Suraishkumar, G. K. (2023). A reactive species reactions module for integration into genome-scale metabolic models for improved insights: Application to cancer. Metabolic Engineering, 80, 78-93. https://doi.org/10.1016/j.ymben.2023.08.006